The sense of taste is responsible for detecting and responding to sweet, bitter, sour, salty and umami (amino acid) stimuli. It is also capable of distinguishing between these various taste modalities to generate innate behavioral responses. For instance, animals are vigorously averse to bitter-tasting compounds, but are attracted to sweet and umami stimuli. To examine taste signal detection and information processing, we have focused on the isolation and characterization of sweet, umami and bitter taste receptors. These receptors provide powerful molecular tools to delineate the organization of the taste system, and to help define the logic of taste coding.
Two families of candidate mammalian taste receptors, the T1Rs and T2Rs, have been implicated in sweet, umami and bitter detection. The T2Rs are a family of ˜30 taste-specific GPCRs distantly related to opsins, and clustered in regions of the genome genetically linked to bitter taste in humans and mice (Adler et al., Cell 100, 693-702 (2000); Matsunami et al., Nature, 404, 601-604 (2000)). Several T2Rs have been shown to function as bitter taste receptors in heterologous expression assays, substantiating their role as bitter sensors (Chandrashekar et al., Cell, 100, 703-711 (2000); Bufe et al., Nat Genet, 32, 397-401 (2002)). Most T2Rs are co-expressed in the same subset of taste receptor cells (Adler, E. et al., Cell 100, 693-702 (2000)), suggesting that these cells function as generalized bitter detectors.
The T1Rs are a small family of 3 GPCRs expressed in taste cells of the tongue and palate epithelium, distantly related to metabotropic glutamate receptors, the calcium sensing receptor and vomeronasal receptors (Hoon et al., Cell., 96, 541-551 (1999); Kitagawa et al., Biochem Biophys Res Commun, 283, 236-242 (2001); Max et al.,| Sac. Nat Genet, 28, 58-63 (2001); Montmayeur et al. Nat Neurosci, 4, 492-498 (2001); Nelson et al., Cell, 106, 381-390 (2001); Sainz et al., J Neurochem, 77, 896-903 (2001)). T1Rs combine to generate at least two heteromeric receptors: T1R1 and T1R3 form an L-amino acid sensor, which in rodents recognizes most amino acids, and T1R2 and T1R3 associate to function as a broadly tuned sweet receptor (Nelson, G. et al., Cell, 106, 381-390 (2001); Nelson, G. et al., Nature, 416, 199-202 (2002); Li, X. et al., Proc Natl Acad Sci USA, 99, 4692-4696 (2002); see also WO 00/06592, WO 00/06593, and WO 03/004992).
Animals can detect a wide range of chemically distinct sweet tasting molecules, including natural sugars, artificial sweeteners, D-amino acids and intensely sweet proteins. How many different receptors does it take to taste the sweet universe? The human and rodent T1R2+3 heteromeric sweet receptors respond in cell-based assays to all classes of sweet compounds, and do so with affinities that approximate their respective in vivo psychophysical and/or behavioral thresholds (Nelson et al., Cell, 106, 381-390 (2001); Li et al., Proc Natl Acad Sci USA, 99, 4692-4696 (2002)). At a fundamental level, the evolution of sweet taste most likely reflects the need to detect and measure sugar content in potential food sources. Therefore, a single broadly tuned receptor for natural sugars might be all that is required. On the other hand, a number of studies with various sugars and artificial sweeteners insinuate the possibility of more than one sweet taste receptor (Schiffman et al., Pharmacol Biochem Behav, 15, 377-388 (1981); Ninomiya et al., J Neurophysiol, 81, 3087-3091 (1999)).
In humans, monosodium L-glutamate (MSG) and L-aspartate, but not other amino acids, elicit a distinctive savory taste sensation called umami (Maga, 1983). Notably, unlike the rodent T1R1+3, the human T1R1+3 amino acid taste receptor is substantially more sensitive to L-glutamate and L-aspartate than to other L-amino acids (Li et al., Proc Natl Acad Sci USA, 99, 4692-4696 (2002)). These findings led to the proposal that T1R1+3 may be the mammalian umami receptor (Nelson. et al., Nature, 416, 199-202 (2002); Li. et al., Proc Natl Acad Sci USA, 99, 4692-4696 (2002)). However, a number of studies, including the recent analysis of T1R3 KO mice (Damak et al., Science, 301, 850-853 (2003)) have suggested that umami taste is mediated by mGluR4t, a truncated variant of the metabotropic glutamate receptor (Chaudhari et al., Neurosci, 16, 3817-3826 (1996); Chaudhari. et al., Nat Neurosci, 3, 113-119 (2000)).
How are the different taste qualities encoded at the taste cell level? In mammals, taste receptor cells are assembled into taste buds that are distributed in different papillae in the tongue epithelium. Each taste bud contains 50-150 cells, including precursor cells, support cells, and taste receptor cells (Lindemann, Physiol Rev, 76, 718-766 (1996)). The receptor cells are innervated by afferent fibers that transmit information to the taste centers of the cortex through synapses in the brain stem and thalamus. In the simplest model of taste coding at the periphery, each taste modality would be encoded by a unique population of cells expressing specific receptors (e.g. sweet cells, bitter cells, salt-sensing cells, etc.). In this scenario, our perception of any one taste quality would result from the activation of distinct cell types in the tongue (labeled line model). Alternatively, individual taste cells could recognize multiple taste modalities, and the ensemble firing pattern of many such broadly tuned receptor cells would encode taste quality (across fiber model).
Recently, we showed that T1Rs and T2Rs are expressed in completely non-overlapping populations of receptor cells in the lingual epithelium (Nelson et al., Cell, 106, 381-390 (2001)), and demonstrated that bitter-receptor expressing cells mediate responses to bitter but not to sweet or amino acid tastants (Zhang et al., Cell, 112, 293-301 (2003)). Together, these results argued that taste receptor cells are not broadly tuned across all modalities, and strongly supported a labeled line model of taste coding at the periphery. A fundamental question we address now is how many types of cells and receptors are necessary to mediate sweet and umami, the two principal attractive taste modalities. We now show that sweet and umami tastes are exclusively mediated by T1Rs, and demonstrate that genetic ablation of individual T1R subunits selectively affects these two attractive taste modalities. The identification of cells and receptors for sweet and umami sensing also allowed us to devise a strategy to separate the role of receptor activation from cell stimulation in encoding taste responses. We show that animals engineered to express a modified k-opioid receptor in T1R2+3-expressing cells become specifically attracted to a k-opioid agonist, and prove that activation of sweet-receptor expressing cells, rather than the T1R receptors themselves, is the key determinant of behavioral attraction to sweet tastants. Finally, we now demonstrate that T1R1 alone, either as a monomer or as a homodimer, acts as a receptor for naturally occurring sugars.